Skip to contentSkip to conversation

linux-acpi-2.6/test on Thinkpad T61: Battery Absent

DISCUSSION Yves-Alexis Perez 3 participants 4 messages Sep 10, 2008 – Sep 11, 2008 ACPI

lore
Share & tools
lore
#0 lore raw

Hi,

I just pulled linux-acpi-2.6/test on my git tree, and rebuild the kernel, and now it doesn't detect any battery:

ACPI: Battery Slot [BAT0] (battery absent)

My tree is at faca89759dd7d4d87309f56cd09ae885a296fb4c and I'll try to bisect.

Cheers,

Signature · 1 line
Yves-Alexis
#1 lore raw
Quoted 9 lines · Yves-Alexis Perez

Hi,

I just pulled linux-acpi-2.6/test on my git tree, and rebuild the kernel, and now it doesn't detect any battery:

ACPI: Battery Slot [BAT0] (battery absent)

My tree is at faca89759dd7d4d87309f56cd09ae885a296fb4c and I'll try to bisect.

Can the battery be detected on the upstream kernel of 2.6.27-rc6? Please attach the output of acpidump, lspci -vxxx and dmesg.(Both dmesg on acpi_test git tree and 2.6.27-rc6 are needed).
Thanks.

Quoted 2 lines

Cheers,

#2 lore raw

Ok, it seems my mail didn't reach the list. In case it was because of attachments, I resend it.

Quoted 1 line · Zhao Yakui

Can the battery be detected on the upstream kernel of 2.6.27-rc6?

Yes.

Quoted 2 lines

Please attach the output of acpidump, lspci -vxxx and dmesg.(Both dmesg on acpi_test git tree and 2.6.27-rc6 are needed).

Ok, files are there: http://molly.corsac.net/~corsac/acpi/

Thanks for any help.

Signature · 1 line
Yves-Alexis
#3 lore raw
In reply to Yves-Alexis Perez
Quoted 2 lines · Yves-Alexis Perez

My tree is at faca89759dd7d4d87309f56cd09ae885a296fb4c and I'll try to bisect.

Ok, playing with git bisect, it seems the offending commit is:

50c0f43ff1199bbea5f0418d40df32f4d4029b13 is first bad commit commit 50c0f43ff1199bbea5f0418d40df32f4d4029b13

Author: Andi Kleen <[email protected]>
Date:   Thu Sep 4 15:13:29 2008 +0200

ACPI: EC: do transaction from interrupt context

From: Alexey Starikovskiy <[email protected]>

It is easier and faster to do transaction directly from interrupt
context
rather than waking control thread.
Also, cleaner GPE storm avoidance is implemented.

AK: Merged with earlier io port patch

:040000 040000 5b8664bf958e6455709348a1a52a14a92dd3119c
7d8ecce115fec5dd1e808fc345b8724def2ad606 M drivers

Not sure exactly what the problem is, though.

Cheers,

Signature · 1 line
Yves-Alexis